An apparatus and method of generating an immersion is provided. The apparatus for generating the immersion includes: a foreground/background separation unit to separate an input image into a foreground area and a background area using lightness information of the input image; an edge classification unit to classify the foreground area into an edge area and a non-edge area using the lightness information; a detail enhancement unit to enhance detail of the edge area of the foreground area; and a blurring performing unit to perform blurring of the non-edge area and the background area.
Legal claims defining the scope of protection, as filed with the USPTO.
1. An apparatus for generating an immersion, the apparatus comprising: a foreground/background separation unit to separate an input image into a foreground area and a background area using lightness information of the input image; an edge classification unit to classify the foreground area into an edge area and a non-edge area using the lightness information; a detail enhancement unit to enhance detail of the edge area of the foreground area; and a blurring performing unit to perform blurring of the non-edge area and the background area.
2. The apparatus of claim 1 , wherein the foreground/background separation unit comprises: an image generation unit to generate an image for emphasizing a high frequency component and an image for eliminating the high frequency component, for the input image using the lightness information of the input image; a local activity calculation unit to calculate a local activity for each of the image of emphasizing the high frequency component and the image for eliminating the high frequency component using a maximum lightness difference value in a block; and a focusing value calculation unit to calculate a focusing value for the input image using the local activity.
3. The apparatus of claim 2 , wherein the foreground/background separation unit separates the input image into the foreground area and the background area by comparing the calculated focusing value and a predetermined threshold of the focusing value.
4. The apparatus of claim 3 , wherein the predetermined threshold of the focusing value is set to maximize a variance between the foreground area and the background area.
5. The apparatus of claim 3 , wherein the foreground/background separation unit separates an area denoting the focusing value greater than or equal to the threshold as the foreground area, and separates an area denoting the focusing value less than the threshold as the background area.
6. The apparatus of claim 1 , wherein the edge classification unit classifies the foreground area into the edge area and the non-edge area using a maximum lightness difference value in a block.
7. The apparatus of claim 1 , wherein the detail enhancement unit enhances a high frequency component of the edge area by adaptively controlling a gain based on a maximum lightness difference value in a block of the edge area.
8. The apparatus of claim 1 , wherein the detail enhancement unit enhances a high frequency component of the edge area using an unsharp masking scheme.
9. The apparatus of claim 1 , wherein the blurring performing unit comprises: a hue value difference calculation unit to calculate a difference between a hue value denoting a peak and a hue value of the background area using a hue histogram of the foreground area; and a blur strength control unit to control a blur strength using the hue value difference and a maximum lightness difference value in a block of the non-edge area.
10. The apparatus of claim 9 , wherein the blur strength control unit increases the blur strength as the hue value difference increases.
11. The apparatus of claim 9 , wherein the blurring performing unit performs bilateral filtering that adaptively eliminates a signal level difference of the edge area and preserves the edge area using the controlled blur strength.
12. A method of generating an immersion, the method comprising: separating an input image into a foreground area and a background area using lightness information of the input image; classifying the foreground area into an edge area and a non-edge area using the lightness information; enhancing detail of the edge area of the foreground area; and performing blurring of the non-edge area and the background area.
13. The method of claim 12 , wherein the separating comprises: generating an image for emphasizing a high frequency component and an image for eliminating the high frequency component, for the input image using the lightness information of the input image; calculating a local activity of each of the image for emphasizing the high frequency component and the image for eliminating the high frequency component using a maximum lightness difference value in a block; and calculating a focusing value for the input image using the local activity.
14. The method of claim 13 , wherein the separating separates the input image into the foreground area and the background area by comparing the calculated focusing value and a predetermined threshold of the focusing value.
15. The method of claim 14 , wherein the predetermined threshold of the focusing value is set to maximize a variance between the foreground area and the background area.
16. The method of claim 14 , wherein the separating separates an area denoting the focusing value greater than or equal to the threshold as the foreground area, and separates an area denoting the focusing value less than the threshold as the background area.
17. The method of claim 12 , wherein the classifying classifies the foreground area into the edge area and the non-edge area using a maximum lightness difference value in a block.
18. The method of claim 12 , wherein the enhancing enhances a high frequency component of the edge area by adaptively controlling a gain based on a maximum lightness difference value in a block of the edge area.
19. The method of claim 12 , wherein the enhancing enhances a high frequency component of the edge area using an unsharp masking scheme.
20. The method of claim 12 , wherein the performing comprises: calculating a difference between a hue value denoting a peak and a hue value of the background area using a hue histogram of the foreground area; and controlling a blur strength using the hue value difference and a maximum lightness difference value in a block of the non-edge area.
21. The method of claim 20 , wherein the controlling increases the blur strength as the hue value difference increases.
22. The method of claim 20 , wherein the performing performs bilateral filtering that adaptively eliminates a signal level difference of the edge area and preserves the edge area using the controlled blur strength.
23. A non-transitory computer-readable recording medium storing a program for implementing a method of generating an immersion, the method comprising: separating an input image into a foreground area and a background area using lightness information of the input image; classifying the foreground area into an edge area and a non-edge area using the lightness information; enhancing detail of the edge area of the foreground area; and performing blurring of the non-edge area and the background area.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 24, 2008
March 13, 2012
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.